www.ProcessEngr.

com

Safety Instrumented Systems (SIS) and Safety Life Cycle

Presented in September 2009

By Jennifer L. Bergstrom Process Engineering Associates, LLC

www.ProcessEngr.com

Safety Instrumented Systems (SIS) and Safety Life Cycle

Agenda: ISA standard that defines Safety Life Cycle Safety concepts (including a lot of new acronyms) Aspects of the Safety Life Cycle and how to take it from cradle to grave Ways to incorporate SIS into process design

www.ProcessEngr.com

ANSI/ISA 84.00.01-2004 for SIS

ANSI/ISA 84.00.01-2004 (IEC 61511-Mod) Application of Safety Instrumented Systems (SIS) for Process Industries :
First version in 1996 Second version approved in 2004 (only addition was Grandfather Clause) OSHA recognizes this standard as a RAGAGEP Defines all steps that encompass the Safety Life Cycle Defines a Safety Instrumented System (SIS)

www.ProcessEngr.com

Safety Life Cycle

Concepts (safety acronyms):
Safety Life Cycle Safety Instrumented System (SIS) Safety Integrity Level (SIL) Safety Instrumented Function (SIF) Safety Requirement Specification (SRS)

www.ProcessEngr.com

Safety Life Cycle

cradle Hazard & Risk Assessment (PHA, SIL Analysis) Design (Execute & Evaluate)

Modification

Installation, Commission, & Validate (FAT, SAT, Functional Proof Test)

Decommission grave

Operations and Maintenance

www.ProcessEngr.com

Safety Life Cycle

Definition: An engineering process designed to achieve a risk-based level of safety with performance criteria that allow versatile technologies and optimal design solutions. -exida In other words, the cycle is meant to guide a safety system from the Risk Assessment cradle to the Decommissioning grave.

www.ProcessEngr.com

Why Safety Life Cycle?

Accidents can and do occur, so in order to help minimize the frequency and/or severity.. Safety Instrumented Systems and Safety Life Cycle were designed to minimize risk

www.ProcessEngr.com

Protection Layers
SIS is used as a protection layer between the hazards of the process and the public (the worse the potential hazard, the more layers required for prevention/protection) Examples:
BPCS (control system), alarms and operator response, SIS, physical devices (PSVs, dikes, flares, deluges, etc.), and other human mitigation (emergency response)

www.ProcessEngr.com

Hazards and Risks in Industry

Risk ups and downs

process risk
P R O C

BPCS alarms SIS mechanical other

Risk

E S S

tolerable

www.ProcessEngr.com

SIF and SIL

Safety Instrumented Function (SIF) is designed to minimize process risks to a tolerable level (or ALARP) Each SIF is assigned a Safety Integrity Level (SIL) during SIL analysis - risk assessment
SIL 0/none lowest risk SIL 1 95% of the SIFs SIL 2 5% of SIFs SIL 3 < 1% (not likely in refineries, but possible in off-shore platforms or nuclear) SIL 4 highest risk (only seen in nuclear industry)

www.ProcessEngr.com

Safety Integrity Level (SIL)

Each SIL rating (increasing in number) must be that much more reliable and available at all times (and costs more for upkeep). Reliability and availability are achieved by:
Design using proper safety components Installation per manufacturers guidelines Testing both at initial startup as well as at specified intervals or after any modification (i.e., via PSSR)

www.ProcessEngr.com

Design
Phase where the SIF/SIS is developed to achieve the risk reduction that is determined in the PHA or SIL Analysis (target SIL). Design options can include: Redundancy (initiators, control system, and/or final elements) Type/style of components (transmitter vs. switch or modulating valve vs. on/off chop valve)
NOTE: If a SIS already exists, then analysis of the existing system is done to determine if the target SIL can be achieved with the current design. (Grandfather Clause)

www.ProcessEngr.com

Design - Type of Failures

When designing or modifying a SIS, keep in mind there are two types of failures:
Safe Failures - FAIL SAFE Dangerous Failures

Safe Failures are the desired failure

Initiated (actual event) Spurious (false undesired but still safe)

Dangerous failures are not desired

Inhibited (bypassed) Dangerous operation (doesnt trip when needed)

www.ProcessEngr.com

Design - Type of Failures

How do we design for safe failures with minimal spurious trips? Voting Logic
Safe 1oo1 1oo2 1oo2D
Best blend of both

Dangerous good best better good better

good good best better best

2oo2 2oo3

(Source: ISA & Exida)

www.ProcessEngr.com

Safety Requirement Specification (SRS)

The design and verification is compiled into a document called the Safety Requirement Specification (SRS)
Information included:
Intent of each SIF (the hazard that is mitigated) Components of each SIF (sensor, logic solver, final element) Calculations to verify the target (required) SIL can be achieved

www.ProcessEngr.com

SIL Verification
SIL verification involves multiple equations to determine the achieved SIL. Some of the components to verify this include:
MTTFS PFD RRF (inverse of PFD or 1/PFD)

NOTE: SIL 1 achieves a RRF of 10 to 100

www.ProcessEngr.com

SIL Verification
If the required SIL can not be achieved with the initial design, some options are:
More frequent proof testing Add redundancy (i.e., initiating device, control system, final element) Install smarter device (i.e., HART smart transmitter or transmitter vs. switch or relay, smart control valve with diagnostics and feedback and position indication vs. basic control valve) Add protection layers (independent)

www.ProcessEngr.com

General Concepts to Remember in Design

Two ways to achieve lower MTTFS (PFD) or higher RRF to achieve the target SIL:
Diagnostics, diagnostics, diagnostics, Redundancy

Instrumentation with diagnostics is the key!

Feedback information can tell you the condition of the instrument and whether it is ill and about to fail

www.ProcessEngr.com

General Concepts to Remember in Design

Transmitter is better than a switch or relay If using switch, solenoid, or relay (anything on/off or discrete), verify that it is normally energized during operation (fail safe) Use dedicated wiring to each device (as much as possible) Minimize common cause failures (i.e., common wires, instrument taps including bridles, or same controller or I/O card) Mechanical devices are the weakest link in the SIF. They can stick if not moved periodically (i.e., PSVs, valves, switches)
To remedy this issue: install double blocks or modulating valves that can be partially stroked

www.ProcessEngr.com

Functional Proof Tests

Proof Tests must be performed at the frequency stated in the SRS to continue the reliability of the SIF. It should include the following information:
Test procedure
Test all bypasses, all individual initiators, and final elements Results of all steps of the procedure Verification that process has been restored to normal operation

Date of test and all personnel performing the test Control logic version # (if available) Results of entire test and any abnormalities found

www.ProcessEngr.com

Final Review
Safety Life Cycle
Guidelines for a safety system from the Risk Assessment cradle to the Decommissioning grave.

SRS
It is only a portion of the Safety Life Cycle, but documents and verifies the SIF design

Employer must also fulfill the SRS timelines as determined in the SRS to the keep the SIF reliable and available to reduce risk.
Functional Proof Test at a specified interval Mission Time replacement interval Document any modifications to SIS or protection layers (MOC)